library(stargazer)
library(car)
library(emmeans)
library(ggplot2)
library(stringr)

load("Data/study_2.RData") #name = data
data$Condition[data$protestIn_speakerOut==1] <- 'Outgroup Speaker'
# ignore warning
data$Condition[data$protestIn_speakerOut==0] <- 'Ingroup Speaker'


#----Descriptive Statistics Tables-----
data_descriptives <- subset(data, select = c(empathic_concern, perspective_taking, protestIn_speakerOut, protestIn_speakerOut_IncCenter,
                                             censorship, punishment, sympathy,schadenfreude
                                             ))
stargazer(as.data.frame(data_descriptives), title="Descriptive Statistics for Full Sample", omit.summary.stat = c("p25", "p75"), out="Tables and Figures/study2_descriptives.tex")
# copy / pasted and appears as Table B1 in paper.

# Estimate interaction effects ------------------------------
hypo3a <- lm(censorship ~ empathic_concern + as.factor(Condition) + empathic_concern*as.factor(Condition), data = data)
hypo4a <- lm(schadenfreude ~ empathic_concern + as.factor(Condition) + empathic_concern*as.factor(Condition), data = data)
extra1a <- lm(punishment ~ empathic_concern + as.factor(Condition) + empathic_concern*as.factor(Condition), data = data)
extra2a <- lm(sympathy ~ empathic_concern + as.factor(Condition) + empathic_concern*as.factor(Condition), data = data)

(mylista <- list(empathic_concern=seq(0,1,by=0.1),Condition=c('Ingroup Speaker','Outgroup Speaker')))
emcontcat  <- emmeans(hypo3a, ~ empathic_concern*Condition, at=mylista)
contrast3a <- contrast(emcontcat, "revpairwise",by="empathic_concern")
emcontcat  <- emmeans(hypo4a, ~ empathic_concern*Condition, at=mylista)
contrast4a <- contrast(emcontcat, "revpairwise",by="empathic_concern")
emcontcat  <- emmeans(extra1a, ~ empathic_concern*Condition, at=mylista)
contrast1a <- contrast(emcontcat, "revpairwise",by="empathic_concern")
emcontcat  <- emmeans(extra2a, ~ empathic_concern*Condition, at=mylista)
contrast2a <- contrast(emcontcat, "revpairwise",by="empathic_concern")

hypo3b <- lm(censorship ~ perspective_taking*Condition, data = data)
hypo4b <- lm(schadenfreude ~ perspective_taking*Condition, data = data)
extra1b <- lm(punishment ~ perspective_taking*Condition, data = data)
extra2b <- lm(sympathy ~ perspective_taking*Condition, data = data)

(mylistb <- list(perspective_taking=seq(0,1,by=0.1),Condition=c('Ingroup Speaker','Outgroup Speaker')))
emcontcat  <- emmeans(hypo3b, ~ perspective_taking*Condition, at=mylistb)
contrast3b <- contrast(emcontcat, "revpairwise",by="perspective_taking")
emcontcat  <- emmeans(hypo4b, ~ perspective_taking*Condition, at=mylistb)
contrast4b <- contrast(emcontcat, "revpairwise",by="perspective_taking")
emcontcat  <- emmeans(extra1b, ~ perspective_taking*Condition, at=mylistb)
contrast1b <- contrast(emcontcat, "revpairwise",by="perspective_taking")
emcontcat  <- emmeans(extra2b, ~ perspective_taking*Condition, at=mylistb)
contrast2b <- contrast(emcontcat, "revpairwise",by="perspective_taking")

stargazer(hypo3a, extra1a, extra2a, hypo4a) # Table B2
stargazer(hypo3b, extra1b, extra2b, hypo4b) # Table B3


# Previous study: SCK and make Figure 3 -------------
load("Data/simas_clifford_kirkland_study2_replication.RData")
colnames(table)[3:9] <- c("punishment", "sympathy", "schadenfreude", "Condition", "empathic_concern", "perspective_taking", "censorship")
table$Condition <- ifelse(table$Condition==1, "Outgroup Speaker",
                          "Ingroup Speaker")

sck_hypo3a <- lm(censorship ~ empathic_concern + as.factor(Condition) + empathic_concern*as.factor(Condition), data = table)
sck_hypo4a <- lm(schadenfreude ~ empathic_concern + as.factor(Condition) + empathic_concern*as.factor(Condition), data = table)
sck_extra1a <- lm(punishment ~ empathic_concern + as.factor(Condition) + empathic_concern*as.factor(Condition), data = table)
sck_extra2a <- lm(sympathy ~ empathic_concern + as.factor(Condition) + empathic_concern*as.factor(Condition), data = table)

(mylista <- list(empathic_concern=seq(0,1,by=0.1),Condition=c('Ingroup Speaker','Outgroup Speaker')))
emcontcat  <- emmeans(sck_hypo3a, ~ empathic_concern*Condition, at=mylista)
sck_contrast3a <- contrast(emcontcat, "revpairwise",by="empathic_concern")
emcontcat  <- emmeans(sck_hypo4a, ~ empathic_concern*Condition, at=mylista)
sck_contrast4a <- contrast(emcontcat, "revpairwise",by="empathic_concern")
emcontcat  <- emmeans(sck_extra1a, ~ empathic_concern*Condition, at=mylista)
sck_contrast1a <- contrast(emcontcat, "revpairwise",by="empathic_concern")
emcontcat  <- emmeans(sck_extra2a, ~ empathic_concern*Condition, at=mylista)
sck_contrast2a <- contrast(emcontcat, "revpairwise",by="empathic_concern")

sck_hypo3b <- lm(censorship ~ perspective_taking*Condition, data = table)
sck_hypo4b <- lm(schadenfreude ~ perspective_taking*Condition, data = table)
sck_extra1b <- lm(punishment ~ perspective_taking*Condition, data = table)
sck_extra2b <- lm(sympathy ~ perspective_taking*Condition, data = table)

(mylistb <- list(perspective_taking=seq(0,1,by=0.1),Condition=c('Ingroup Speaker','Outgroup Speaker')))
emcontcat  <- emmeans(sck_hypo3b, ~ perspective_taking*Condition, at=mylistb)
sck_contrast3b <- contrast(emcontcat, "revpairwise",by="perspective_taking")
emcontcat  <- emmeans(sck_hypo4b, ~ perspective_taking*Condition, at=mylistb)
sck_contrast4b <- contrast(emcontcat, "revpairwise",by="perspective_taking")
emcontcat  <- emmeans(sck_extra1b, ~ perspective_taking*Condition, at=mylistb)
sck_contrast1b <- contrast(emcontcat, "revpairwise",by="perspective_taking")
emcontcat  <- emmeans(sck_extra2b, ~ perspective_taking*Condition, at=mylistb)
sck_contrast2b <- contrast(emcontcat, "revpairwise",by="perspective_taking")

output.list <- c("contrast1a", "contrast2a", "contrast3a", "contrast4a",
                 "contrast1b", "contrast2b", "contrast3b", "contrast4b",
                 "sck_contrast1a", "sck_contrast2a", "sck_contrast3a", "sck_contrast4a",
                 "sck_contrast1b", "sck_contrast2b", "sck_contrast3b", "sck_contrast4b")

output <- lapply(output.list, function(x){
                                          model <- x
                                          x <- get(x) 
                                          x <- data.frame(x)
                                          x$model <- model 
                                          return(x)}
                                          )
output_ec <- do.call("rbind", output[c(1:4,9:12)])
output_pt <- do.call("rbind", output[c(5:8,13:16)])

output_ec$analysis <- ifelse(str_detect(output_ec$model, "sck")==TRUE, "SCK", "GSR")
output_ec$dv <- ifelse(str_detect(output_ec$model,"1")==TRUE, "Punishment",
                ifelse(str_detect(output_ec$model,"2")==TRUE, "Sympathy",
                ifelse(str_detect(output_ec$model,"3")==TRUE, "Censorship","Schadenfreude")))
output_ec$lo <- output_ec$estimate - 1.96*output_ec$SE
output_ec$hi <- output_ec$estimate + 1.96*output_ec$SE

output_pt$analysis <- ifelse(str_detect(output_pt$model, "sck")==TRUE, "SCK", "GSR")
output_pt$dv <- ifelse(str_detect(output_pt$model,"1")==TRUE, "Punishment",
                ifelse(str_detect(output_pt$model,"2")==TRUE, "Sympathy",
                ifelse(str_detect(output_pt$model,"3")==TRUE, "Censorship","Schadenfreude")))
output_pt$lo <- output_pt$estimate - 1.96*output_pt$SE
output_pt$hi <- output_pt$estimate + 1.96*output_pt$SE

cbp2 <- c("#000000", "#E69F00", "#56B4E9", "#009E73",
          "#F0E442", "#0072B2", "#D55E00", "#CC79A7")

ec <- ggplot(output_ec, aes(x=empathic_concern, y=estimate, fill=analysis)) +
  geom_line() + 
  geom_ribbon(aes(ymin=lo, ymax=hi), alpha=.5) + 
  facet_wrap(~dv, nrow=2, scales="free") +
  geom_hline(yintercept=0, linetype="dashed") +
  scale_fill_manual(values = cbp2) +
  xlab("Empathic Concern")+
  theme(legend.title=element_blank())


pt <- ggplot(output_pt, aes(x=perspective_taking, y=estimate, fill=analysis)) +
  geom_line() + 
  geom_ribbon(aes(ymin=lo, ymax=hi), alpha=.5) + 
  facet_wrap(~dv, nrow=2, scales="free") +
  geom_hline(yintercept=0, linetype="dashed") +
  scale_fill_manual(values = cbp2) +
  xlab("Perspective taking") +
  theme(legend.title=element_blank())

ec_pt_plot <- ggpubr::ggarrange(ec, pt, nrow=2, common.legend=TRUE, legend="bottom")

# Make smaller version for main paper
selection <- c("Censorship", "Schadenfreude")
colnames(output_ec)[2] <- "z"
colnames(output_pt)[2] <- "z"
output_ec$z.var <- "EC"
output_pt$z.var <- "PT"
output_ec_pt <- dplyr::bind_rows(output_ec[output_ec$dv%in%selection,], output_pt[output_pt$dv%in%selection,])

ec_pt <- ggplot(output_ec_pt, aes(x=z, y=estimate, fill=analysis)) +
  geom_line() + 
  geom_ribbon(aes(ymin=lo, ymax=hi), alpha=.5) + 
  facet_grid(z.var~dv, scales="free") +
  geom_hline(yintercept=0, linetype="dashed") +
  scale_fill_manual(values = cbp2) +
  xlab("Moderator") +
  theme(legend.title=element_blank(),
        legend.position="bottom")


ggsave(ec_pt,file="Tables and Figures/ec_pt_plot_small.jpg", dpi=1000, height=3, width=3)
# This is Figure 3 in the paper



# Robustness models not reported in the paper-----
# table(data$left_right_placement_1)
# data$left_right_extremism <- (data$left_right_placement_1-5)^2
# censorship <- lm(censorship ~ empathic_concern + protestOut_speakerIn + empathic_concern*protestOut_speakerIn +
#                    left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                    empathic_concern*ClosenessProtestGroup_01 + protestOut_speakerIn*ClosenessProtestGroup_01
#                    , data = data)
# summary(censorship)
# punishment <- lm(punishment ~ empathic_concern + protestOut_speakerIn + empathic_concern*protestOut_speakerIn +
#                    left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                    empathic_concern*ClosenessProtestGroup_01 + protestOut_speakerIn*ClosenessProtestGroup_01
#                  , data = data)
# summary(punishment)
# sympathy <- lm(sympathy ~ empathic_concern + protestOut_speakerIn + empathic_concern*protestOut_speakerIn +
#                  left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                  empathic_concern*ClosenessProtestGroup_01 + protestOut_speakerIn*ClosenessProtestGroup_01
#                , data = data)
# summary(sympathy)
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestOut_speakerIn + empathic_concern*protestOut_speakerIn +
#                       left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                       empathic_concern*ClosenessProtestGroup_01 + protestOut_speakerIn*ClosenessProtestGroup_01
#                     , data = data)
# summary(schadenfreude)
# stargazer::stargazer(censorship,punishment,sympathy,schadenfreude)
# 
# censorship <- lm(censorship ~ perspective_taking + protestOut_speakerIn + perspective_taking*protestOut_speakerIn +
#                    left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                    perspective_taking*ClosenessProtestGroup_01 + protestOut_speakerIn*ClosenessProtestGroup_01
#                  , data = data)
# summary(censorship)
# punishment <- lm(punishment ~ perspective_taking + protestOut_speakerIn + perspective_taking*protestOut_speakerIn +
#                    left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                    perspective_taking*ClosenessProtestGroup_01 + protestOut_speakerIn*ClosenessProtestGroup_01
#                  , data = data)
# summary(punishment)
# sympathy <- lm(sympathy ~ perspective_taking + protestOut_speakerIn + perspective_taking*protestOut_speakerIn +
#                  left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                  perspective_taking*ClosenessProtestGroup_01 + protestOut_speakerIn*ClosenessProtestGroup_01
#                , data = data)
# summary(sympathy)
# schadenfreude <- lm(schadenfreude ~ perspective_taking + protestOut_speakerIn + perspective_taking*protestOut_speakerIn +
#                       left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                       perspective_taking*ClosenessProtestGroup_01 + protestOut_speakerIn*ClosenessProtestGroup_01
#                     , data = data)
# summary(schadenfreude)
# stargazer::stargazer(censorship,punishment,sympathy,schadenfreude)
# 
# censorship <- lm(censorship ~ empathic_concern*perspective_taking*protestOut_speakerIn +
#                    left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                    perspective_taking*ClosenessProtestGroup_01 + empathic_concern*ClosenessProtestGroup_01 +
#                    protestOut_speakerIn*ClosenessProtestGroup_01
#                  , data = data)
# summary(censorship)
# punishment <- lm(punishment ~ empathic_concern*perspective_taking*protestOut_speakerIn +
#                    left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                    perspective_taking*ClosenessProtestGroup_01 + empathic_concern*ClosenessProtestGroup_01 +
#                    protestOut_speakerIn*ClosenessProtestGroup_01
#                  , data = data)
# summary(punishment)
# sympathy <- lm(sympathy ~ empathic_concern*perspective_taking*protestOut_speakerIn +
#                  left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                  perspective_taking*ClosenessProtestGroup_01 + empathic_concern*ClosenessProtestGroup_01 +
#                  protestOut_speakerIn*ClosenessProtestGroup_01
#                , data = data)
# summary(sympathy)
# schadenfreude <- lm(schadenfreude ~ empathic_concern*perspective_taking*protestOut_speakerIn +
#                       left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 +
#                       perspective_taking*ClosenessProtestGroup_01 + empathic_concern*ClosenessProtestGroup_01 +
#                       protestOut_speakerIn*ClosenessProtestGroup_01
#                     , data = data)
# summary(schadenfreude)
# stargazer::stargazer(censorship,punishment,sympathy,schadenfreude)
# 
# censorship <- lm(censorship ~ empathic_concern*protestL_speakerR +
#                    left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 + ClosenessProtestGroup_01
#                  , data = data)
# summary(censorship)
# punishment <- lm(punishment ~ empathic_concern*protestL_speakerR +
#                    left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 + ClosenessProtestGroup_01
#                  , data = data)
# summary(punishment)
# sympathy <- lm(sympathy ~ empathic_concern*protestL_speakerR +
#                  left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 + ClosenessProtestGroup_01
#                , data = data)
# summary(sympathy)
# schadenfreude <- lm(schadenfreude ~ empathic_concern*protestL_speakerR +
#                       left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 + ClosenessProtestGroup_01
#                     , data = data)
# summary(schadenfreude)
# stargazer::stargazer(censorship,punishment,sympathy,schadenfreude)
# 
# 
# censorship <- lm(censorship ~ perspective_taking*protestL_speakerR +
#                    left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 + ClosenessProtestGroup_01
#                  , data = data)
# summary(censorship)
# punishment <- lm(punishment ~ perspective_taking*protestL_speakerR +
#                    left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 + ClosenessProtestGroup_01
#                  , data = data)
# summary(punishment)
# sympathy <- lm(sympathy ~ perspective_taking*protestL_speakerR +
#                  left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 + ClosenessProtestGroup_01
#                , data = data)
# summary(sympathy)
# schadenfreude <- lm(schadenfreude ~ perspective_taking*protestL_speakerR +
#                       left_right_placement_1 + left_right_extremism + Perceived_UNRealism01 + ClosenessProtestGroup_01
#                     , data = data)
# summary(schadenfreude)
# stargazer::stargazer(censorship,punishment,sympathy,schadenfreude)
# 
# 
# 
# #hypo 3A: Exploratory --------
# censorship1 <- lm(censorship ~ empathic_concern + protestL_speakerR + empathic_concern*protestL_speakerR, data = data)
# summary(censorship1)
# 
# punishment <- lm(punishment ~ empathic_concern + protestL_speakerR + empathic_concern*protestL_speakerR, data = data)
# summary(punishment)
# 
# sympathy <- lm(sympathy ~ empathic_concern + protestL_speakerR + empathic_concern*protestL_speakerR, data = data)
# summary(sympathy)
# 
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestL_speakerR + empathic_concern*protestL_speakerR, data = data)
# summary(schadenfreude)
# 
# (mylist <- list(empathic_concern=seq(0.0,1,by=0.01),protestL_speakerR=c(0,1)))
# emmip(censorship1,protestL_speakerR~empathic_concern,at=mylist, CIs=TRUE)
# 
# subset(data, !is.na(protestL_speakerR)) %>% 
#   ggplot() +
#   aes(x = empathic_concern, y = censorship, group = protestL_speakerR, color = protestL_speakerR) +
#   geom_smooth(method = "lm")  +
#   labs(x = "Empathic Concern") +
#   labs(y = "Desire to Censor the Speaker")
# 
# subset(data, !is.na(protestL_speakerR)) %>% 
#   ggplot() +
#   aes(x = empathic_concern, y = punishment, group = protestL_speakerR, color = protestL_speakerR) +
#   geom_smooth(method = "lm")  +
#   labs(x = "Empathic Concern") +
#   labs(y = "Desire to Censor the Speaker")
# 
# 
# #hypo 4A: Exploratory
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestOut_speakerIn + empathic_concern*protestOut_speakerIn, data = data)
# summary(schadenfreude)
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestOut_speakerIn + empathic_concern*protestOut_speakerIn +
#                       Left_leaning + ClosenessProtestGroup_01 + Perceived_UNRealism01 + 
#                       Language_EN + UvA
#                       , data = data)
# summary(schadenfreude)
# 
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestOut_speakerIn + empathic_concern*protestOut_speakerIn +
#                       Left_leaning + ClosenessProtestGroup_01 + Perceived_UNRealism01 + 
#                       Language_EN + UvA +
#                       empathic_concern*ClosenessProtestGroup_01
#                     , data = data)
# summary(schadenfreude)
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestIn_speakerOut + empathic_concern*protestIn_speakerOut +
#                       Left_leaning + ClosenessProtestGroup_01 + Perceived_UNRealism01 + 
#                       Language_EN + UvA +
#                       empathic_concern*ClosenessProtestGroup_01
#                     , data = data)
# summary(schadenfreude)
# 
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestOut_speakerIn + empathic_concern*protestOut_speakerIn +
#                       Left_leaning + ClosenessProtestGroup_01 + Perceived_UNRealism01 + 
#                       Language_EN + UvA +
#                       empathic_concern*ClosenessProtestGroup_01+ ClosenessProtestGroup_01*protestOut_speakerIn
#                     , data = data)
# summary(schadenfreude)
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestIn_speakerOut + empathic_concern*protestIn_speakerOut +
#                       Left_leaning + ClosenessProtestGroup_01 + Perceived_UNRealism01 + 
#                       Language_EN + UvA +
#                       empathic_concern*ClosenessProtestGroup_01 + ClosenessProtestGroup_01*protestIn_speakerOut
#                     , data = data)
# summary(schadenfreude)
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestIn_speakerOut + empathic_concern*protestIn_speakerOut +
#                       Left_leaning + DistanceProtest_01 + Perceived_UNRealism01 + 
#                       Language_EN + UvA +
#                       empathic_concern*DistanceProtest_01 + DistanceProtest_01*protestIn_speakerOut
#                     , data = data)
# summary(schadenfreude)
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestOut_speakerIn + empathic_concern*protestOut_speakerIn +
#                       Left_leaning + DistanceProtest_01 + Perceived_UNRealism01 + 
#                       Language_EN + UvA +
#                       empathic_concern*DistanceProtest_01 + DistanceProtest_01*protestOut_speakerIn
#                     , data = data)
# summary(schadenfreude)
# 
# schadenfreude1 <- lm(schadenfreude ~ empathic_concern + protestOut_speakerIn +
#                       ClosenessProtestGroup_01 + Perceived_UNRealism01 + Left_leaning + 
#                       protestOut_speakerIn*empathic_concern*ClosenessProtestGroup_01
#                     , data = data)
# summary(schadenfreude1)
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestIn_speakerOut +
#                       Left_leaning + ClosenessProtestGroup_01 + Perceived_UNRealism01 +  
#                       empathic_concern*ClosenessProtestGroup_01*protestIn_speakerOut
#                     , data = data)
# summary(schadenfreude)
# schadenfreude <- lm(schadenfreude ~ empathic_concern + protestIn_speakerOut +
#                       Left_leaning + DistanceProtest_01 + Perceived_UNRealism01 +  
#                       empathic_concern*DistanceProtest_01*protestIn_speakerOut
#                     , data = data)
# summary(schadenfreude)
# 
# CP_Plus2SD <- 1
# CP_MEAN <- round(mean(data$ClosenessProtestGroup_01,na.rm=T),2)
# CP_Minus2SD <- round(mean(data$ClosenessProtestGroup_01,na.rm=T) - 2*sd(data$ClosenessProtestGroup_01,na.rm=T),2)
# 
# (mylist <- list(empathic_concern=seq(0.25,1,by=0.01),ClosenessProtestGroup_01=c(CP_Plus2SD,CP_MEAN,CP_Minus2SD)))
# emmip(schadenfreude1,ClosenessProtestGroup_01~empathic_concern,at=mylist, CIs=TRUE)
# 
# (mylist <- list(empathic_concern=seq(0.25,1,by=0.01),ClosenessProtestGroup_01=c(0,1)))
# emmip(schadenfreude1,ClosenessProtestGroup_01~empathic_concern,at=mylist, CIs=TRUE)
# 
# hypo1a.red <- lm(censorship ~ empathic_concern + protestIn_speakerOut_IncCenter + empathic_concern*protestIn_speakerOut_IncCenter
#              , data = data, subset=left_right_placement_1%in%c(0:3,7:10))
# 
# hypo2a.red <- lm(punishment ~ empathic_concern + protestIn_speakerOut + empathic_concern*protestIn_speakerOut
#              , data = data, subset=left_right_placement_1%in%c(0:3,7:10))
# hypo3a.red <- lm(sympathy ~ empathic_concern + protestIn_speakerOut + empathic_concern*protestIn_speakerOut
#              , data = data, subset=left_right_placement_1%in%c(0:3,7:10))
# summary(hypo3a)
# hypo4a.red <- lm(schadenfreude ~ empathic_concern + protestIn_speakerOut + empathic_concern*protestIn_speakerOut
#              , data = data, subset=left_right_placement_1%in%c(0:3,7:10))
# 
# 
# 
# 
# 
# 
# 
# 
